Alterations in the expression of certain midgut genes of Spodoptera littoralis (Boisd.) (Lepidoptera: Noctuidae) larvae and midgut histopathology in response to Bacillus thuringiensis Cry1C toxin

Background: Bacillus thuringiensis (Bt) utilization as a biological control agent is highly recommended due to its safety, specificity, and efficiency. Importance of the entomocidal Cry proteins secreted by Bt is dramatically increased subsequent Cry genes transformation into a number of economic crops, rendering them protection against insect attack. In the last decade, insect resistance against transgenic Bt crops is gradually raised in several lepidopteran pests. A better understanding of the processing of Bt Cry1C toxin in the larval midgut of the lepidopteran pest species, the cotton leaf worm, Spodoptera littoralis (Boisd.), is very important to characterize the main regulatory elements of Bt tolerance. Results: The present study aimed to define factors that are involved in insect tolerance toward Bt Cry1C through evaluating the mRNA level of trypsin (Try), aminopeptidase N (APN), alkaline phosphatase (ALP), cadherin (Cad), and cytochrome P450 (CYP) in both susceptible and cry1C tolerant strains of S. littoralis. Total RNAs were extracted from susceptible and tolerant strains to construct cDNAs. Quantitative real-time polymerase chain reaction (qPCR) showed a significant upregulation of CYP gene in tolerant strain. In contrast, the levels of expression of Try, ALP, and Cad were significantly downregulated in tolerant strain. APN relative mRNA expression did not show significant differences between susceptible and tolerant strains. Histologically, the midgut of late third-instar larvae of tolerant population S. littoralis showed vacuolization of the epithelium and disruption of both the peritrophic membrane and the striated boarder compared to the susceptible strain. Conclusions: Obtained data indicated a relationship between exposing to Bt Cry1C toxin and alteration of CYP, Try, ALP, and Cad expression in midgut of S. littoralis. These results may be an evidence for the important roles of CYP, Try, ALP, and Cad in the resistance development and toxicity to Bt Cry1C. The results are useful for further illustrating of Bt Cry1C processing and S. littoralis tolerance.


Background
The Egyptian cotton leaf worm, Spodoptera littoralis (Boisd.) (Lepidoptera: Noctuidae), is a polyphagous pest in subtropical and tropical regions. The long-term applications of conventional pesticides to control pests, including S. littoralis, have led to the developments of resistance, negative impacts on the ecosystems, and unhealthy hazards to human and natural enemies. These problems are incentives to search for alternative safe and effective control measures with different modes of action. Biological control agents can be used where chemical pesticides are banned or where pests have developed resistance to standard chemicals and play an important role in integrated pest management (IPM) programs (Abdelhadi et al. 2016 andvan Lenteren et al. 2018). Among these approaches is the use of the entomopathogenic bacterium, Bacillus thuringiensis (Bt) that provides a valuable alternative to chemical insecticides. (Dingha et al. 2004). Bt synthesizes delta-endotoxin protein crystals (Cry) that belong to a class of bacterial toxins are known as pore-forming toxins that specifically affect cell permeability and disrupting cellular integrity of insectmidgut cells. Cry toxins have a potent and specific insecticidal activity and can kill their insect larval hosts through a complex multi-step process resulting in the formation of a pore in the membrane of midgut epithelial cells (Soberón et al. 2018). Differences in the degree of solubilization possibly cause variability in the level of toxicity among Cry proteins. Solubilization is followed by proteolytic activation of the Cry protoxin by midgut digestive enzymes; the activated toxin crosses the peritrophic membrane, reaches the brush border membrane vesicles (BBMVs) of the midgut epithelium, and binds to the primary receptor cadherin in the microvilli (Pigott and Ellar 2007), which undergoes proteolytic cleavages that induce toxin oligomerization. After oligomers insertion, sequentially osmotic lysis of midgut epithelial cells occurs, followed by the creation of cytolytic pores in the membrane of midgut epithelial cells, septicemia, and ultimately insect death (Pardo-López et al. 2013).
Development of transgenic crops that express cry toxins was a remarkable breakthrough in pest control instead of the use of chemical insecticides (ISAAA 2017). The expression of certain Cry toxins in transgenic crops providing a more targeted and effective way to control insect pests in agriculture (James 2009). The expression of Cry protein in transgenic plants protects the insecticidal toxin from UV degradation and precisely targets chewing and boring insects. Nevertheless, the continuous expression of cry toxins in transgenic crops results in a gradual resistance development (Xiao and Wu 2019). Extensive studies have established that alteration of genes for toxins activation, and toxin-binding receptors confers high levels of Bt resistance in insects and subsequently threatens the sustained success of transgenic Bt crops (Fabrick et al. 2019). During attempts to unwind the complexity of Bt toxicity, other mechanisms associated with Cry toxins mode of action such as detoxication enzyme activities or innate immunity response were proposed (Boyer et al. 2012). Recently, alteration in CYP expression and activity due to exposure to cry toxins has been reported (Dhania et al. 2019). Thus, the correlation between CYP enzymatic activity and mechanism of Bt resistance is implied (Shabbir et al. 2019). A better understanding of the mode of action of Bt toxins and the mechanism of resistance at the molecular level is essential to improve strategies for monitoring and coping with the development of insect resistance, which assists in designing insect control strategies and extending the efficacy of Bt toxins as a control agent (Jin et al. 2018). Due to the existence of limited genetic information for S. littoralis, thus, the present study aimed to elucidate the molecular mechanism of Bt Cry1C tolerance, as a step to delay resistance development and guarantee a long-term efficacy of Bt in biological control.

Insect culture
Two strains of S. littoralis larvae were used in the present study: a susceptible strain and a Cry1C-tolerant strain. A susceptible strain was provided by Insect Biotechnology and Molecular Biology Unit, Plant Protection Research Institute, Agricultural Research Center (ARC), Egypt, and fed on castor bean leaves, Ricinus communis, under the laboratory conditions (25 ± 1°C, 70-80% RH and 14-h light:10-h dark photoperiod) for several years according to El-Defrawi et al. (1964). A Cry1C-tolerant strain was generated by selection pressure of a fieldstrain collected from Qalyubia Governorate, Egypt. Selection pressure was carried out by exposing newly hatched neonates to a semi-artificial diet (Rajagopal et al. 2009), mixed with partially purified Bt Cry1C toxin isolated from Bt entomocidus strain. Bt entomocidus strain was provided by Agricultural Genetic Engineering Research Institute, ARC, Egypt. The toxin dose increased gradually with the successive generations depending on the mortality percentage in each generation. Survived larvae were then shifted to castor bean leaves until pupation and adult emergence (Moussa et al. 2020). This process was repeated every generation to rise up Cry1Ctolerant strain. Ten generations were accomplished.

Bt Cry1C toxin preparation
The methods described by Moussa et al. (2016) were followed for Bt Cry1C toxin purification. Briefly, the bacterial cells were cultured for 3 to 5 days in T3 medium (per liter: 1.5 g yeast extract, 0.005 g of MnCl 2 , 0.05 M phosphate buffer pH 6.8, 3.0 g tryptone, and 2.0 g tryptose) at 30°C/150 rpm. The spores and crystals were harvested at 5500 rpm/10 min at 4°C and then washed six times with 50 mM EDTA at 9500 rpm/10 min at 4°C. The pellet was resuspended in 50 mM Tris HCl, 5 mM EDTA (pH 7), and preserved in −20°C until solubilization. Resuspension was centrifuged at 9500 rpm/10 min at 4°C and supernatant was removed. Two milliliters of solubilization buffer (50 mM Na 2 CO 3 , 10 mM DTT, pH 10.5) was added and incubated for 4 h at 37°C/200 rpm, and then centrifuged at 14,000 rpm/30 min at 4°C. Supernatant was aliquoted and kept at −20°C for further use. Toxin concentration was obtained by Bradford method (Bradford, 1976) and toxin integrity was checked on 10% SDS PAGE.

Bioassay
Newly hatched neonates of a susceptible and a Cry1Ctolerant strains of S. littoralis were fed separately on a semi-artificial diet mixed with five concentrations of purified Bt Cry1C toxin: 0.2, 0.4, 0.8, 1.6, and 3.2 μg/g for susceptible strain and 2.0, 4.0, 8.0, 16.0, and 32.0 μg/ g for Cry1C-tolerant strain, following the techniques described by Rajagopal et al. (2009) and Moussa et al. (2016). Mortality percentage was recorded daily till the 7th day. Each concentration was replicated 3 times with 10 larvae each. A parallel control of 10 untreated larvae was also run. Each control was replicated 3 times. Treatments were conducted at different times, using different larval batches. Mortality percentage of each treatment was corrected using Abbott's formula (Abbott 1925). The 50% lethal concentration (LC 50 ) for each strain was estimated with the probit analysis. Resistance ratio (RR) was calculated as the LC 50 for the tolerant strain divided by the LC 50 for the susceptible strain.

Histopathological studies
Biopsy samples of the middle portion of the midgut of Cry1C-tolerant strain of S. littoralis late third-instar larvae were taken. Parallel controls of untreated susceptible strain larvae were also run. For light microscopy, the midgut was fixed in Bouin's solution. After dehydration in a graded ethanol series, the midgut was embedded in paraffin wax and cut 5 μ thick using a rotary microtome. The sections were stained with hematoxylin and eosin and photographed with an Axiophot (Zeiss) light microscope according to Bancroft and Gamble (2008).

RNA extraction
The midguts of newly molted 3rd instar larvae of susceptible and Cry1C-tolerant strains of S. littoralis were dissected prior to RNA extraction. Triple biological replicates were conducted with five larval midguts each. Gene JET RNA Purification Kit (cat # K0731) was used for total RNA extraction according to the manufacturer's instructions, and was quantified by absorbance at 260 nm. One percentage agarose gel electrophoresis was utilized for RNA integrity determination. Extracted RNAs were treated with DNase I, RNase-free (cat # EN0521), then reverse transcribed using an oligo (dT) 15 primer with GoScript TM Reverse Transcription System (cat # A5000) as per manufacturer's instructions.
Quantitative real-time polymerase chain reaction qPCR was conducted on the Stratagene Mx3005P QPCR System (Agilent Technologies Germany GmbH & Co.KG, Waldbronn, Germany) using the SybrGreen method with Maxima SYBR Green qPCR Master Mix (2×) (cat # K0251). Primers were designed based on conserved regions of the targeted genes isolated from other lepidopteran insects and deposited in the GenBank database (Table 1). MultAlin (http://multalin.toulouse.inra. fr/multalin/) was used for nucleotide sequence alignment for each gene (Corpet 1988). Primers were designed using GenScript Primer Design tool and expected to amplify about 100-150 bp fragment (Table 2). A 28S rRNA gene served as a reference gene. The selected genes were amplified under the following conditions: 95°C for 10 min, followed by forty cycles of 95°C for 30 s, 53°C for 60 s, and 72°C for 30 s. The melting curve analysis was utilized to analyze the specificity of the qPCR product. Relative expression fold changes were calculated by using formula 2 −ΔΔCT , which was proposed by Livak and Schmittgen (2001). The comparative Ct (ΔΔCt) was measured by subtracting ΔCt of calibrator from ΔCt of treated samples.

Statistical analysis
EPA Probit analysis program (version 1.5) (kindly provided by Dr. Gujar G.T., New Delhi, India) was utilized to estimate the 50% lethal concentration (LC 50 ), 95% fiducial limits (FL), and the slope for results of the bioassay by probit analysis. The real-time polymerase chain reaction was done in three wells (replicates) for all genes. To determine the significance (P ≤ 0.05) among the mean differences of the groups, the independent unpaired Student's t test was used the Statistical Package for the Social Sciences version 23 (SPSS, IBM, Armonk, NY, USA). All data were analyzed with a significance level of 5%.

Results
Tolerance level of S. littoralis larvae to Bt Cry1C toxin In this study, the field S. littoralis strain underwent selection pressure for 10 generations to rise up Cry1C-tolerant strain. Bt Cry1C toxin was partially purified and used for S. littoralis treatment. The SDS-PAGE showed the presence of the Bt Cry1C toxin at 135 kDa protein (Fig. 1). After 10 generations, the tolerant strain showed 76.67% mortality rate at 32.0 μg toxin per gram diet, while it was 70.00% at 3.2 μg/g in susceptible strain (Table 3). As a result of the subsequent selection pressure for 10 generations, the LC 50 of the tolerant strain reached 12.263 μg/gm compared to 1.895 μg/gm LC 50 of the susceptible strain. Therefore, the resistance ratio reached up to 6.5-fold the susceptible strain (Table 4). The difference between the LC 50 values of susceptible and tolerant strains was significant (P < 0.05), where the respective 95% fiducial limits were not overlapped.

Histopathological studies
Briefly, the midgut of the 3rd instar larvae of S. littoralis is made up of a single layer of epithelium resting on a basement membrane, surrounded by a layer of circular muscle fibers and an outer longitudinal muscle coat. The epithelium consists of 3 types of cells: columnar cells, goblet cells, and regenerative cells. The apical surface of each columnar cell bordering with the gut lumen is covered with microvilli. The gut lumen is lined with the peritrophic membrane (Fig. 2a). Light microscopy revealed that the midgut of late 3rd instar larvae of Cry1C-tolerant population of S. littoralis showed vacuolization of the epithelium and disruption of both the peritrophic membrane and the striated boarder (Fig. 2b). The lumen was collapsed and globular bodies and cytoplasmic fragments were observed pinching off from the tip of some of the epithelial cells within the lumen vicinal to the deteriorated peritrophic membrane.

qPCR gene expression
Understanding the mode of action of Bt toxins and the mechanism of tolerance is a critical point to identify the effective way to use these toxins in S. littoralis control.
In order to investigate genes that might be involved in the insect's tolerance, a total RNAs were isolated from both susceptible and tolerant strains. The cDNAs were constructed, and qPCR was applied to assess the relative transcript abundance of cadherin (Cad), alkaline phosphatase (ALP), trypsin (Try), aminopeptidase N (APN), and cytochrome P450 (CYP) in susceptible and tolerant strains. The results reported here provided evidence about midgut receptors, proteases, and detoxification enzymes that associated with the processing of Bt Cry1C toxin. qPCR results showed about (72%) reduction (P < 0.05) of Cad transcripts in tolerant strain than the susceptible one (Fig. 3). Abundance of ALP as a secondary receptor promoting the localization of toxin in the midgut was evaluated. Reduction of ALP transcripts in tolerant strain compared to susceptible strain was detected. In this study, APN transcript abundance did not differ significantly between susceptible and tolerant strains. About 49% reduction of Try transcripts in tolerant strain was revealed. Cytochrome P450 (CYP6AB14) was significantly over-transcribed in tolerant strain ( Fig. 3 and Table 5).

Discussion
Bacillus thuringiensis (Bt), a soil bacterium, is the extreme successfully utilized biopesticide in agriculture. Bt contained insecticidal protein genes that are primary utilized for insect control in transgenic crops. Insect resistance to Bt, it was a great challenge to sustainable success of the most extensively utilized transgenic crops (Tabashnik et al. 2013). There were many types of proteins that present in the midgut epithelial cells were either reported as transporters, which facilitate the toxicity functions or digestive proteases. For example, Cad (Walsh et al. 2018) and ALPs (Ren et al. 2018) that were shown to interact with Cry toxin(s) produced by different Bt strains and were described as Cry toxin functional receptors in midgut epithelium of insects. Bt was known to cause changes in the cell membrane, a common appearance of midgut degeneration (Cavados et al. 2004). In the present study, tolerant strain showed some disruptions in the gut lumen. However, of these disruptions, the gut functions were not affected and larvae can survive. This may indicated the regeneration of the tissues affected by Cry1C toxin. Forcada et al. (1999) proposed that enhanced gut healing response was a mechanism that involves Bt resistance. Castagnola and Jurat-Fuentes (2016) suggested that the increased production of new midgut growth factors cause an enhanced midgut regenerative response in resistant insects. This may be another way for the insect to tolerate Bt toxicity.
The results reported in the present study showed also the downregulation of cadherin (Cad) gene in tolerant strain than in the susceptible one. Downregulation of Cad and the subsequent reduction of Bt Cry1C binding and toxin oligomerization indicated the essential role of Cad gene in toxicity of Bt Cry1C toxin and tolerance selection. Recent studies have reported that at least two receptors on the insect midgut membrane interact with Bt toxins (Pigott and Ellar 2007). The midgut Cad (the first receptor) binds to activate Bt toxins with high affinity, and the interaction with Cad helped oligomerization of the toxins through a proteolytic process (Soberón et al. 2009). Therefore, the Cad gene low expression could lead to loose of oligomerization of the Bt Cry1C toxin, which in turn reflected the toxicity process. This result  was consistent with earlier studies. For instances, downregulation of Cad has been reported in lepidopteran species including Helicoverpa armigera (Hübner) (Wang et al. 2005), Diatraea saccharalis (Fabricius) (Yang et al. 2011), Ostrinia furnacalis (Guenée) (Jin et al. 2014), Pectinophora gossypiella (Saunders) (Fabrick et al. 2019), and also in dipteran species as Aedes aegypti (Linnaeus) (Bonin et al. 2009). Literature proposed that downregulation of Cad was associated with Cry toxin resistance. addition, the results showed that aminopeptidase N (APN) transcript abundance did not differ significantly between susceptible and tolerant strains. This may refer that APN did not associate with Bt Cry1C tolerance in S. littoralis. In contrast, previous studies showed that alterations in expression of APN have been associated with Bt resistance in several species of insect pests (Zhang et al. 2017). The Bt oligomers had a high binding affinity to APN or alkaline phosphatase (ALP) (a secondary receptor), which finally led to oligomers insertion into the midgut cell membrane, with resulting cell lysis. It also has been proposed that binding of Bt toxins to the Cad may activate a cellular signaling pathway leading to cell death without the involvement of APN (Zhang et al. 2006). This may explain obtained results and supports the unchanged expression level of APN between the two strains. Thus, APN might not be a core receptor for Bt Cry1C in S. littoralis and did not involve in toxicity. Further experiments are needed to test this hypothesis.
In the Bt Cry1C-tolerant strain, ALP expression level was significantly reduced than the susceptible one. This lack of expression showed the crucial role of ALP in toxin insertion through the midgut and pore formation. The role of ALP in the susceptibility of insects to Cry toxins has been demonstrated in several studies Adang, 2004 andQiu et al. 2018). Knockdown of ALP gene in the rice borer, Chilo suppressalis, exhibited decreased susceptibility to transgenic Cry1A rice (Qiu et al. 2018). Also, other transcripts, which were downregulated in this study in the tolerant strain that was previously documented as remarkable Bt genes/proteins involved in insecticide resistance in many insects, which is trypsin-like serine protease. Thus, downregulation of trypsin will be resulted in improper and insufficient activation of Bt Cry1C protoxin, which may be a key factor in tolerance development. Cis-mutations identified in the promoter region of a trypsin gene conferred high resistance to Cry1Ac in H. armigera (Liu et al. 2014). However, the most reports indicated that variations in toxin activation were commonly not a main mechanism of resistance to Bt proteins (Wei et al. 2016). Cytochrome P450 (CYP6AB14) was significantly overtranscribed in the present investigation. This induction may reveal the relation between detoxification activity and tolerance. Expression and activity of P450 enzyme was different through resistant insect populations, some studies reported upregulation and others reported downregulation (Vellichirammal et al. 2015). In previous studies, P450 gene was reported to confer resistance and was involved in detoxifications of xenobiotics (Pavlidi et al. 2018), as well as trypsin, which is considered the essential protein involved in Bt toxin activation and detoxification (Liu et al. 2014). Therefore, further studies are needed to be conducted to identify the major and precise role of cytochrome P450 in detoxification and tolerance of Bt.

Conclusion
The present study concluded that toxin activation, binding, and detoxification were critical procedures of Bt Cry1C toxification in S. littoralis. Trypsin, aminopeptidase N, alkaline phosphatase, cadherin, and cytochrome P450 are among different proteins that involved in Bt resistance and toxicity. Further studies are required to better understand how these proteins are regulated and their role in Bt tolerance.  ΔCt of Tol.